Machine Learning for Healthcare
Ali Akbar Septiandri
@aliakbars
Outline
What is machine learning?
How to identify cats or dogs in an image?
Identifying Cats or Dogs
Model
Image Processing
Feature Extraction
y = σ(β0 + β1x1 + β2x2 + β3x3)
Logit model from defined features
“Fundamentally, machine learning involves building mathematical models to help understand data.”
…but it might not even work!
Enter: Deep Learning
(Artificial) Neural Networks ~ Deep Learning
It’s that simple*!
Output
Feature extraction
+ model training
Input
“No free lunch”
Opportunities
Verbal screening using mobile app
Stanford ML Group
“...it could diagnose 14 diseases with 80% accuracy–in other words, about as well as a real radiologist.”
Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening (Wu et al., 2019)
Making out of focus microscopy images in focus again
Early Warning System
System for Electronic Notification and Documentation (SEND) by Sensyne Health
Natural Language Processing (NLP) for clinical notes
DeepMind’s AlphaFold
“...diagnosing and treating diseases believed to be caused by misfolded proteins, such as Alzheimer’s, Parkinson’s, Huntington’s and cystic fibrosis.”
Challenges
Accuracy
“Correlation does not imply causation” | Image from xkcd
Understanding Causality
Graphical presentation of confounding in directed acyclic graphs (Suttorp et al., 2014)
Missing Data
Adversarial Attacks
Bias in ML
“...a project to look for skin cancer in photographs. It turns out that dermatologists often put rulers in photos of skin cancer, for scale, but that the example photos of healthy skin do not contain rulers. To the system, the rulers (or rather, the pixels that we see as a ruler) were just differences between the example sets, and sometimes more prominent than the small blotches on the skin. So, the system that was built to detect skin cancer was, sometimes, detecting rulers instead.” (Evans, 2019)
Future of ML4HC
Optimal treatment strategies, e.g. for sepsis (Komorowski et al., 2018)
Where are we now?
Thank you.
aliakbars@live.com
@__aliakbars__
http://uai.aliakbars.com